The present invention relates to a method for bit error rate measurements in a cell-based telecommunication system.
In cell-based telecommunication systems, the information is transferred by means of cells of fixed or variable byte-length. These cells typically have an overhead section wherein control, management and routing information is embedded and a payload section, carrying user information.
Cell-based telecommunication systems comprise for example ATM (Asynchronous Transfer Mode) communication networks. ATM cells for instance have a fixed length of 53 bytes, 8 bytes of which constitute the overhead section or the so called ATM cell header, reserved for overhead information.
The whole transmitted cell-stream in a cell-based network is generally composed of two different types of cells: on the one hand used cells carrying in their payload section the proper user information and on the other hand unused cells with no user information in their payload section. The unused cells are generated and inserted amid the used cells to be transmitted in order to provide a continuous cell-stream between transmitting means and receiving means.
Bit errors which occur during transmission of these cells, have a negative impact onto the quality of the connection. A bit error which occurs in the payload section of the cell implies a falsification of the transmitted information; wherein a bit error which occurs in the overhead section of the cell potentially implies a falsification of the destination address, and thus a loss of the cell.
Therefore in cell-based telecommunication systems, the bit error rates (ratio of bits transferred erroneously due to noise or impairments of the physical transmission medium in proportion to the totality of transferred bits) are traditionally measured during initialisation of the system or during a test phase. Hereby a predetermined signal is sent from a transmitter to a receiver. The arriving signal is then analysed at the receiver""s side by comparing the transmitted signal with a generated reference signal.
In an ADSL (Asymmetric Digital Subscriber Line) system for instance, the signal to noise ratio (SNR) is measured at the different carrier frequencies during initialisation of the system. This signal to noise ratio can be seen as a measure for the expected bit error rate. The results of these measurements afterwards are used to determine the bit allocations, i.e. the way wherein data bits are distributed over the different carriers that constitute a DMT (Discrete Multi Tone) symbol, to cope with the maximum allowable bit error rate of 10xe2x88x927 prescribed by the ADSL standard. The SNR measurements and the use thereof in the bit allocation process are described in the ANSI (American National Standards Institute) ADSL Standard T1E1.4, in paragraph 12 entitled xe2x80x98Initializationxe2x80x99 on pages 87-111 and in paragraph 6.5 entitled xe2x80x98Tone Orderingxe2x80x99 on Pages 36-37.
Traditional bit error rate measurement methods have the disadvantage, that normal operation of the system has to be interrupted and the system has to be brought in a test phase. During this test phase, no transmission of proper user information is possible.
The object of the present invention is to provide a method for bit error rate measurements in a cell-based telecommunication system during normal operation of the system.
According to the invention, this problem is solved by a method for bit error rate measurements in a cell-based telecommunication system in accordance to claim 1.
The method for bit error rate measurements in a cell-based telecommunication system, comprises the steps of generating a first bit pattern at a first location of the telecommunication system, transmitting the first bit pattern from the first location to a second location of the telecommunication system, thereby obtaining a transmitted bit pattern, generating a second bit pattern at the second location, the second bit pattern being identical with the first bit pattern and comparing the transmitted bit pattern with the second bit pattern. According to the invention the first bit pattern is transmitted in at least one unused cell, the unused cell being generated during normal operation of the telecommunication system in the case of lack of used cells, carrying user information, to be transmitted. The number of discrepancies counted during the comparison serves as a measure for the bit error rate.
It has to be noted, that in this context, the expression xe2x80x9cunused cellxe2x80x9d means cells, which do not carry user data in their payload section. In other words xe2x80x9cunused cellxe2x80x9d can either designate an idle cell or an unassigned ATM cell.
One essential advantage of the invention, compared with conventional methods is that the bit error rate measurements are performed during normal operation of the telecommunication system. Accordingly the telecommunication system has not to be interrupted and brought in a test phase in which no transmission of proper user information is possible.
In a cell-based telecommunication system, e.g. an ATM-network, where unused cells, carrying no user information, are generally inserted amid the used cells in order to provide a continuous cell-stream between transmitting means and receiving means, bit error rate measurements according to the present method advantageously need no additional bandwidth by using those unused cells.
It has to be noted, that the generated bit patterns are independent from any transmitted user information and that the bit pattern generated at the first location and the reference bit pattern generated at the second location are identical. This means, that the number of detected discrepancies between the transmitted bit pattern and the second bit pattern really represents a measure for the bit error rate at the moment of transmission, because the bit errors only can occur in the test bit pattern itself. This is not the case in systems, where control information, calculated on the basis of user data to be transmitted, is transmitted together with said user information and the transmitted control information is compared to control information calculated on the basis of the transmitted user data. In fact, in these systems, bit errors can occur as well in the transmitted user data as in the transmitted control information, thus leading to a possible falsification of the measurements results.
In a preferred embodiment, the steps for bit error rate measurements according to the present method are repeated at each generation of an unused cell. The accuracy of the achieved bit error rates is increasing with the number of repetitions, i.e. with the number of comparisons between transmitted bit patterns and generated bit patterns at the receiver.
The comparison between the transmitted bit pattern and the second bit pattern is preferably bit synchronised, which enables a fast evaluation of bit error rates during measurements.
The generation of the second bit pattern may be synchronised with the arrival of the transmitted bit pattern at a comparator means for comparing said transmitted bit pattern with said second bit pattern.
The first bit pattern is advantageously a predetermined bit sequence or an algorithm-based counter bit-sequence. Both types of bit patterns may be easily reproduced at the receiving means.
The present invention also relates to a device for bit error rate measurement in a cell-based telecommunication system. A device according to the invention comprises a transmitter, a receiver, a first bit pattern generator for generating a first bit pattern to be transmitted from the transmitter to the receiver in order to obtain a transmitted bit pattern, a second bit pattern generator for generating a second bit pattern at the receiving means, the second bit pattern being identical to the first bit pattern, a comparator for comparing the transmitted bit pattern with the second bit pattern and means for inserting the first bit pattern in at least one unused cell, said unused cell being generated during normal operation of said telecommunication system in the case of lack of used cells to be transmitted.
In a preferred embodiment, the device for bit error rate measurements comprises a synchroniser for activating the second bit pattern generator at each arrival of an unused cell at said comparator.
In the case of a bi-directional communication system, each subscriber station is simultaneously sending and receiving cells, thus provided with transceivers, each transceiver comprising transmitting means and receiving means.